Sabermetricians commonly preach regression to the mean, and to that end FIP is used as a common indicator of regression candidates. Some prefer xFIP because that stat attempts to ‘normalize’ home run rates, and that’s fine. I view each as tools that are useful and different, rather than better or worse. Today we’ll look at the data from 2016 and 2017 to see just how often the ‘regression monster’ successfully pulls the ERA in FIP’s direction.
Pitcher
|
2016 ERA | 2016 FIP | 2016 FIP-2016 ERA | 2017 ERA | 2017 ERA-2016 ERA |
Brandon Finnegan | 3.98 | 5.19 | 1.21 | 4.15 | 0.17 |
Dan Straily | 3.76 | 4.88 | 1.12 | 4.26 | 0.50 |
Kyle Hendricks | 2.13 | 3.20 | 1.07 | 3.03 | 0.90 |
Ian Kennedy | 3.68 | 4.67 | 0.99 | 5.38 | 1.70 |
Jon Lester | 2.44 | 3.41 | 0.97 | 4.33 | 1.89 |
Tanner Roark | 2.83 | 3.79 | 0.96 | 4.67 | 1.84 |
Felix Hernandez | 3.82 | 4.63 | 0.81 | 4.36 | 0.54 |
JA Happ | 3.18 | 3.96 | 0.78 | 3.53 | 0.35 |
Michael Fulmer | 3.06 | 3.76 | 0.70 | 3.83 | 0.77 |
Chase Anderson | 4.39 | 5.09 | 0.70 | 2.74 | -1.65 |
Marco Estrada | 3.48 | 4.15 | 0.67 | 4.98 | 1.50 |
Cole Hamels | 3.32 | 3.98 | 0.66 | 4.20 | 0.88 |
Jason Hammel | 3.83 | 4.48 | 0.65 | 5.29 | 1.46 |
Jake Odorizzi | 3.69 | 4.31 | 0.62 | 4.14 | 0.45 |
Hector Santiago | 4.70 | 5.31 | 0.61 | 5.63 | 0.93 |
Listed above are the 15 pitchers with the greatest positive differential between their 2016 FIP and their 2016 ERA, limited to pitchers with 150+ IP two seasons passed. In theory, these were the ‘luckiest’ pitchers of 2016 and we should have expected a worse performance in 2017. The average FIP-ERA differential among these pitchers in 2016 was 0.83, while the average differential between 2017 ERA and 2016 ERA was 0.82. As for the latter, that number was skewed by the anomaly in Milwaukee. Chase Anderson was the ONLY pitcher among these fifteen regression candidates who improved last year. An increased strand rate and an average fastball up 2 mph were part of that surprise package. Throwing Anderson’s numbers out, the average increase in ERA among the rest was roughly 1 full run.
Pitcher | 2017 ERA | 2017 FIP | 2017 FIP-ERA | 2018 NFBC ADP |
Lance Lynn | 3.43 | 4.82 | 1.39 | 226 |
Jose Urena | 3.82 | 5.20 | 1.38 | 436 |
Andrew Cashner | 3.40 | 4.61 | 1.21 | 572 |
Ervin Santana | 3.28 | 4.46 | 1.18 | 211 |
Gio Gonzalez | 2.96 | 3.93 | 0.97 | 149 |
Robbie Ray | 2.89 | 3.72 | 0.83 | 43 |
Marcus Stroman | 3.09 | 3.90 | 0.81 | 123 |
Clayton Kershaw | 2.31 | 3.07 | 0.76 | 6 |
John Lackey | 4.59 | 5.30 | 0.71 | 467 |
Jake Arrieta | 3.53 | 4.16 | 0.63 | 94 |
Ariel Miranda | 5.12 | 5.73 | 0.61 | 563 |
Alex Wood | 2.72 | 3.32 | 0.60 | 87 |
Taijuan Walker | 3.49 | 4.04 | 0.55 | 214 |
Drew Pomeranz | 3.32 | 3.84 | 0.52 | 215 |
Jason Vargas | 4.16 | 4.67 | 0.51 | 495 |
I will be underweight in terms of my ownership of all the above pitchers this year. In the past, I’ve tried to kick against the goads, because there are exceptions to the rule. You can find reasons to believe that FIP’s gravity won’t have an effect on a given pitcher. Last year that temptation was Kyle Hendricks, with the ultra high spin rate on his changeup (not to mention two type of changeups), and the low spin rate on his two-seamer inducing ground balls I was seduced into trusting him in 2017. This year some might be allured by Byron Buxton’s defense in centerfield playing into Ervin Santana’s FIP. Hands off for me this year. This doesn’t mean to put these guys on a ‘do not draft’ list, but I’d recommend expecting something closer to 2017 FIP than their 2017 ERA.
Pitcher | 2017 ERA | 2017 FIP | 2017 ERA- 2017 FIP | 2018 NFBC ADP |
Jason Hammel | 5.29 | 4.37 | 0.92 | 599 |
Jordan Zimmermann | 6.08 | 5.18 | 0.90 | 656 |
Jeff Samardzija | 4.42 | 3.61 | 0.81 | 138 |
Matt Moore | 5.52 | 4.75 | 0.77 | 481 |
Chris Archer | 4.07 | 3.40 | 0.67 | 46 |
Clayton Richard | 4.79 | 4.23 | 0.56 | 540 |
Tanner Roark | 4.67 | 4.13 | 0.54 | 258 |
Michael Wacha | 4.13 | 3.63 | 0.50 | 234 |
Jose Quintana | 4.15 | 3.68 | 0.47 | 66 |
Mike Foltynewicz | 4.79 | 4.33 | 0.46 | 359 |
Chris Sale | 2.90 | 2.45 | 0.45 | 11 |
Jimmy Nelson | 3.49 | 3.05 | 0.44 | 382 |
Masahiro Tanaka | 4.74 | 4.34 | 0.40 | 89 |
Marco Estrada | 4.98 | 4.61 | 0.37 | 330 |
Ty Blach | 4.78 | 4.42 | 0.36 | 832 |
This year FIP is going to help us a lot more in identifying pitchers to avoid than signaling which ones to target. Jeff Samardzija should improve considerably, but the market has already priced that in to his average draft position. For the third straight season, Chris Archer’s ERA has underperformed FIP. So don’t bet too heavily on a ‘correction.’ Jimmy Nelson recently received good news and started his throwing program earlier than expected, but there is still no timetable for his return and shoulder injuries are unpredictable. Tanner Roark should improve and rack up the wins in Washington. Jose Quintana is very interesting moving to the National League, increasing his strikeouts and lowering his ERA, and joining the Cubbies should help the win total. Finally, those wondering if Chris Sale’s elite performance in Fenway Park was for real, it was.
Follow Greg @Liquidhippo.